OBJECTIVE The aim of this study is to assess the performance of deep learning convolutional neural networks (CNNs) in segmenting gadolinium-enhancing lesions using a large cohort of multiple sclerosis (MS) patients. METHODS A three-dimensional (3D) CNN model was trained for segmentation of gadolinium-enhancing lesions using multispectral magnetic resonance imaging data (MRI) from 1006 relapsing-remitting MS patients. The network performance was evaluated for three combinations of multispectral MRI used as input: (U5) fluid-attenuated inversion recovery (FLAIR), T2-weighted, proton density-weighted, and pre- and post-contrast T1-weighted images; (U2) pre- and post-contrast T1-weighted images; and (U1) only post-contrast T1-weighted images. Segmentation performance was evaluated using the Dice similarity coefficient (DSC) and lesion-wise true-positive (TPR) and false-positive (FPR) rates. Performance was also evaluated as a function of enhancing lesion volume. RESULTS The DSC/TPR/FPR values averaged over all the enhancing lesion sizes were 0.77/0.90/0.23 using the U5 model. These values for the largest enhancement volumes (>500 mm3) were 0.81/0.97/0.04. For U2, the average DSC/TPR/FPR values were 0.72/0.86/0.31. Comparable performance was observed with U1. For all types of input, the network performance degraded with decreased enhancement size. CONCLUSION Excellent segmentation of enhancing lesions was observed for enhancement volume ⩾70 mm3. The best performance was achieved when the input included all five multispectral image sets.

中文翻译：

---

多发性硬化症中钆增强病变的深度学习分割


目的 本研究的目的是使用大量多发性硬化症 (MS) 患者来评估深度学习卷积神经网络 (CNN) 在分割钆增强病变方面的性能。方法 使用来自 1006 名复发缓解型多发性硬化症患者的多光谱磁共振成像 (MRI) 数据，训练三维 (3D) CNN 模型来分割钆增强病变。使用多光谱 MRI 的三种组合作为输入来评估网络性能：(U5) 流体衰减反转恢复 (FLAIR)、T2 加权、质子密度加权以及对比前和对比后 T1 加权图像； (U2) 对比前和对比后 T1 加权图像； (U1) 仅对比后 T1 加权图像。使用 Dice 相似系数 (DSC) 和病变真阳性 (TPR) 和假阳性 (FPR) 率评估分割性能。还根据增强病变体积的函数来评估性能。结果 使用 U5 模型，所有增强病灶大小的 DSC/TPR/FPR 平均值为 0.77/0.90/0.23。最大增强体积 (>500 mm3) 的这些值为 0.81/0.97/0.04。对于 U2，平均 DSC/TPR/FPR 值为 0.72/0.86/0.31。在 U1 中观察到了类似的性能。对于所有类型的输入，网络性能随着增强大小的减小而下降。结论 对于增强体积⩾70 mm3，观察到增强病灶的良好分割。当输入包含所有五个多光谱图像集时，可以获得最佳性能。